Descriptive Statistics & Hypothesis Testing
The purpose of these is to transform raw scores into a common unit of measurement.
What are z scores?
Degrees of freedom is directly related to this.
What is the sample size?
The endpoints of the range for a correlation coefficient
What is -1 to 1
Tests conducted after a significant F statistics for an ANOVA.
What are post hoc tests and/or tests of effect size?
This measure of effect size calculates the size of the effect in terms of standard deviation units.
What is Cohen’s d?
This type of measurement is involved when you are measuring a person’s age (years).
What is a ratio scale?
The shape that a t distribution resembles as N increases.
What is the normal distribution (or z distribution)?
The approximate value of a correlation if there is no relationship between the variables.
What is 0?
The approximate value of your obtained F statistic if there is NO effect of your treatment.
What is 1 (or close to 1)?
This measure of effect size is used to estimate a range in which the population parameter is likely derived from the sample statistic.
What is a confidence interval?
The line in the center of a box-and-whisker plot
What is the median?
The standard deviation of the sampling distribution of the differences between the means.
What is the standard error of mean differences?
As one variable increases another variable decreases.
What is a negative or inverse relationship?
Since everyone within a particular group receives the same treatment, this is what within-variability is affected by in an experiment.
What is error (or chance)?
The symbol for the coefficient of determination.
What is r2?
This is what is minimized when power is maximized. (Hint: 1-power = ?)
What is beta or the probability of making a type II error?
The degrees of freedom for a paired-sample t-test with the results, t(79) = 1.98, p < .05 (one-tailed).
What is 78?
The assumption specific to Pearson correlation which requires looking at the shape of a scatterplot.
What is linearity?
The top of the F ratio.
What is between-group variability, MSbetween, or treatment plus error?
This is the critical assumption for all inferential tests we have discussed.
What is independence of observation?
These are the six steps of hypothesis testing in order.
The approximate value of the obtained t statistic if there is NO effect of your treatment.
What is 0 (or close to 0)?
Steps needed to calculate a confidence interval.
What is (1) transform to z, (2) calculate interval, (3) transform to r.
If n = 10, k = 4, and N = 40, what is the value of dfTreatment?
What is 3?
This is what you can conclude about effect size from a significant statistical test.
What is nothing?